ZipRecruiter
Senior AI Performance Engineer
ZipRecruiter, San Francisco, California, United States, 94199
Job DescriptionJob Description
We are Genmo, a research lab dedicated to building open, state-of-the-art models for video towards unlocking the right brain of AGI. Join us in shaping the future of AI and pushing the boundaries of what's possible in video .
Role overview:
As a Deep Learning Performance Engineer at Genmo, you will play a critical role in optimizing the performance of our large generative AI models. Your expertise will ensure that our models run efficiently on clusters, leveraging advanced techniques and tools to enhance their performance. This role is perfect for someone with a deep understanding of deep learning performance bottlenecks, kernel optimization, and distributed training strategies.
Key responsibilities:
Analyze and optimize the performance of massively parallel and distributed systems
Implement and fine-tune distributed training strategies for multi-GPU and multi-node environments
Implement high-performance CUDA, Triton, C++ and PyTorch code.
Profile model performance and identify bottlenecks using tools like NVIDIA NSight Systems, PyTorch Profiler, and TensorFlow Profiler
Develop and maintain benchmarking suites for continuous performance monitoring
Qualifications:
Master's or PhD in Computer Science, Electrical Engineering, or a related field
5+ years of experience in optimizing deep learning models, preferably in a production environment
Must have
Strong programming skills in Python and C++. Experience in training large models using Python & PyTorch and/or TensorFlow including their distributed training frameworks.
Proven track record of optimizing large-scale models (10B+ parameters)
Deep understanding of GPU architecture and CUDA programming
Experience in entire development pipeline from data processing, preparation & data loading to training and inference.
Experience optimizing and deploying inference workloads for throughput and latency across the stack (inputs, model inference, outputs, parallel processing etc.)
Demonstrated expertise in high-performance computing using NVIDIA Triton and CUDA
Demonstrated ability to significantly improve model inference and training speeds through low-level optimizations
Ideal candidates will have:
Knowledge of distributed inference systems for handling high-volume workloads
Strong background in linear algebra, optimization, and machine learning algorithms
Experience with generative AI models (GANs, Diffusion Models, Transformers)
Knowledge of hardware-aware neural architecture design
Experience with high-performance computing (HPC) environments
Contributions to relevant open source projects or research publications
Genmo is an Equal Opportunity Employer. Candidates are evaluated without regard to , , , , , , , , veteran status, or any other characteristic protected by federal or state law. Genmo, Inc. is an E-Verify company and you may review the Notice of E-Verify Participation and the Right to Work posters in English and Spanish.
We are Genmo, a research lab dedicated to building open, state-of-the-art models for video towards unlocking the right brain of AGI. Join us in shaping the future of AI and pushing the boundaries of what's possible in video .
Role overview:
As a Deep Learning Performance Engineer at Genmo, you will play a critical role in optimizing the performance of our large generative AI models. Your expertise will ensure that our models run efficiently on clusters, leveraging advanced techniques and tools to enhance their performance. This role is perfect for someone with a deep understanding of deep learning performance bottlenecks, kernel optimization, and distributed training strategies.
Key responsibilities:
Analyze and optimize the performance of massively parallel and distributed systems
Implement and fine-tune distributed training strategies for multi-GPU and multi-node environments
Implement high-performance CUDA, Triton, C++ and PyTorch code.
Profile model performance and identify bottlenecks using tools like NVIDIA NSight Systems, PyTorch Profiler, and TensorFlow Profiler
Develop and maintain benchmarking suites for continuous performance monitoring
Qualifications:
Master's or PhD in Computer Science, Electrical Engineering, or a related field
5+ years of experience in optimizing deep learning models, preferably in a production environment
Must have
Strong programming skills in Python and C++. Experience in training large models using Python & PyTorch and/or TensorFlow including their distributed training frameworks.
Proven track record of optimizing large-scale models (10B+ parameters)
Deep understanding of GPU architecture and CUDA programming
Experience in entire development pipeline from data processing, preparation & data loading to training and inference.
Experience optimizing and deploying inference workloads for throughput and latency across the stack (inputs, model inference, outputs, parallel processing etc.)
Demonstrated expertise in high-performance computing using NVIDIA Triton and CUDA
Demonstrated ability to significantly improve model inference and training speeds through low-level optimizations
Ideal candidates will have:
Knowledge of distributed inference systems for handling high-volume workloads
Strong background in linear algebra, optimization, and machine learning algorithms
Experience with generative AI models (GANs, Diffusion Models, Transformers)
Knowledge of hardware-aware neural architecture design
Experience with high-performance computing (HPC) environments
Contributions to relevant open source projects or research publications
Genmo is an Equal Opportunity Employer. Candidates are evaluated without regard to , , , , , , , , veteran status, or any other characteristic protected by federal or state law. Genmo, Inc. is an E-Verify company and you may review the Notice of E-Verify Participation and the Right to Work posters in English and Spanish.